VR game interaction delay reduction method based on 5G
By dynamically adjusting packet priority and resource allocation, combined with convolutional neural network acceleration processing, the problem of cross-level delay segmentation in multiplayer real-time competitive VR games is solved, achieving a low-latency and highly smooth interactive experience.
Patent Information
- Application Number
- CN202510847474.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-24
- Publication Date
- 2025-09-16
AI Technical Summary
In multiplayer real-time competitive VR games, the delayed segmentation of the network transmission layer, computing processing layer, and display rendering layer leads to excessively long interaction delays. Especially in high-concurrency environments, the existing network architecture cannot effectively resolve the collaborative contradictions of cross-layer delay segmentation.
By obtaining the packet type identifier, dynamically adjusting the network transmission queue, using convolutional neural networks to accelerate computing processing, interpolation algorithms to smooth display rendering, and combining reinforcement learning to optimize resource scheduling, we ensure priority processing and synchronous feedback of key data packets.
It significantly reduces the delay of aiming operations in VR games, improves the smoothness and accuracy of interaction, improves the player's immersion and gaming experience, and adapts to different network environments and game scenarios.
Smart Images
Figure CN120643904A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of information and communication technologies, and in particular to a method for reducing interaction delay in VR games based on 5G. Background Art
[0002] The rapid development of 5G technology, with its low latency and high bandwidth, has brought new development opportunities to virtual reality (VR) gaming. However, in multiplayer, real-time competitive VR gaming, reducing interactive latency remains a core technical challenge that needs to be addressed. In this scenario, players use gestures to aim and fire weapons, and the system manages latency by breaking it down into the network transmission layer, the computing processing layer, and the display rendering layer.
[0003] The network transport layer is responsible for uploading player gesture data packets from the terminal to the cloud server. Despite the high bandwidth of 5G networks, in high-concurrency environments, when multiple players initiate intensive interactions simultaneously, the network transport layer's packet prioritization becomes a bottleneck. Critical packets (such as aiming commands) compete with less important packets (such as ambient light and shadow synchronization) in the queue, making it impossible for the transport layer to dynamically distinguish which packets are critical for immediate aiming feedback.
[0004] After receiving data packets, the computational processing layer is delayed in processing some key gestures due to server resource scheduling not fully matching the transport layer's priorities, resulting in delays in aiming trajectory generation. Even if the display rendering layer operates efficiently, discontinuities in upstream data input can cause screen updates to not align with player expectations. This synergistic conflict in cross-layer latency division means that while theoretically controllable sub-module latency can actually amplify overall interaction latency due to a lack of a global dynamic tuning mechanism. This technical issue is particularly acute in extreme scenarios with intensive player interaction.
[0005] In addition, although existing network architectures and technologies, such as FlexE, AVB / TSN, and DetNet, can achieve end-to-end deterministic latency transmission to a certain extent, most of these methods are based on new network architectures and cannot be directly implemented in traditional Ethernet. Summary of the Invention
[0006] The present invention provides a method for reducing VR game interaction delay based on 5G, which mainly includes: Obtain the type identifier of each data packet in the network transport layer, determine whether it is a key data packet or a secondary data packet based on a preset threshold, and obtain a priority allocation table; According to the priority allocation table, the network transmission queue is dynamically adjusted in a high-concurrency environment. If a data packet is identified as a critical type, its transmission order is immediately promoted and the adjusted transmission sequence is determined; Extract key data packets from the adjusted transmission sequence and transmit them to the computing processing layer. Then, the corresponding gesture parsing model is matched with the pre-established mapping rules to obtain the initial aiming trajectory data. Obtain the current resource scheduling status in the computing processing layer. If the resource usage exceeds the preset threshold when the processing request of the critical data packet arrives, the dynamic allocation of the backup computing node is triggered and the optimized resource allocation plan is determined; According to the optimized resource allocation plan, the initial aiming trajectory data is input into the assigned computing nodes, and the data is accelerated and processed by the convolutional neural network algorithm to generate accurate aiming trajectory data; Extract the timestamp sequence from the precise aiming trajectory data and determine whether the data generation interval is less than the refresh period of the display rendering layer. If it is greater, interpolation algorithm is used to supplement the intermediate frame data to obtain a continuous trajectory data set. Based on the continuous trajectory dataset, a real-time rendering pipeline is used to convert the data into visual feedback frames in the display rendering layer. The time difference between the feedback frame sequence and the player's gesture input is determined to be below a preset threshold, and a synchronized and optimized frame sequence is obtained. Obtain the synchronized and optimized frame sequence, update the screen content through the refresh mechanism of the head-mounted display device, and determine the final screen output data; The player's expected matching index is extracted from the final screen output data. If the matching degree is lower than the preset threshold, the priority allocation table and resource scheduling plan are adjusted through the reinforcement learning algorithm to generate the optimization parameters for the next round.
[0007] The beneficial effects of the present invention are: This invention effectively resolves bottlenecks at the network transmission layer in high-concurrency environments through precise data packet priority classification and dynamic scheduling. It can quickly distinguish between critical and less important data packets and dynamically adjust the transmission order to ensure the rapid transmission of critical instructions. Furthermore, at the computational processing layer, the invention introduces dynamic allocation of spare computing nodes and a convolutional neural network algorithm to accelerate gesture parsing and aiming trajectory generation, significantly reducing computational latency and improving processing efficiency. At the display rendering layer, this invention uses interpolation algorithms and synchronization optimization mechanisms to supplement intermediate frame data and smooth trajectories, ensuring the consistency of visual feedback and high synchronization with player gestures. Furthermore, optimizations to the real-time rendering pipeline and dynamic refresh mechanism further enhance the player's visual experience. Through a global dynamic tuning mechanism, this invention utilizes reinforcement learning algorithms to continuously optimize priority allocation tables and resource scheduling schemes, resolving the synergistic contradictions of cross-level delay segmentation and ensuring the system maintains low latency and high fluency even in extreme operating scenarios. The multi-level optimization scheme of this invention significantly reduces the latency of aiming operations in VR games, improves the smoothness and accuracy of interactions, and significantly enhances players' immersion and gaming experience. Its dynamic optimization mechanism and modular design enable it to adapt to different network environments and gaming scenarios, demonstrating its wide applicability and scalability. In summary, this invention provides a reliable technical guarantee for the smooth operation of multiplayer real-time competitive VR games, promoting the development of the VR gaming industry. BRIEF DESCRIPTION OF THE DRAWINGS
[0008] For better understanding and implementation, the technical solution of the present application is described in detail below with reference to the accompanying drawings.
[0009] Figure 1 This is a flowchart of a method for reducing 5G-based VR game interaction delay in the present invention. DETAILED DESCRIPTION
[0010] To further illustrate the technical means and effectiveness of the present invention in achieving its intended purpose, exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all possible embodiments consistent with the present application. Rather, they are merely examples of methods and systems consistent with certain aspects of the present application, as detailed in the appended claims.
[0011] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. As used in this application and the appended claims, the singular forms "a," "an," "the," and "the" are intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.
[0012] The following describes in detail the specific implementation methods, features and effects of the present invention in conjunction with the accompanying drawings and preferred embodiments.
[0013] like Figure 1 As shown, a method for reducing VR game interaction delay based on 5G in this embodiment may specifically include: S101. Obtain a type identifier of each data packet in a network transport layer, determine whether it is a key data packet or a secondary data packet based on a preset threshold, and obtain a priority allocation table.
[0014] Obtain the type identification of each data packet from the network transport layer, extract the identification field using data capture technology, determine the original type information of the data packet, and obtain a type identification set; For the type identification set, a preset threshold parameter is obtained, and by comparing the type identification value with the preset threshold, each data packet is judged to be a key data packet or a secondary data packet, and a preliminary classification result is obtained; Based on the preliminary classification results, the support vector machine algorithm is used to extract the characteristics of the data packet type. The classification boundary is trained through the feature vector to determine the precise division between key data packets and secondary data packets, and the optimized classification set is obtained. Extract the ratio of key data packets to secondary data packets from the optimized classification set, calculate the priority weight through statistical analysis, determine the initial value of priority allocation, and obtain a weight allocation table; According to the weight allocation table, the real-time traffic characteristics of the data packets are obtained. If the traffic characteristics exceed the preset dynamic threshold, the priority value of the key data packet is adjusted to obtain a dynamic priority adjustment table; According to the dynamic priority adjustment table, the decision tree algorithm is used to perform secondary verification on the data packet classification. The stability of the priority allocation is judged by the splitting conditions of the tree nodes to obtain the final priority allocation table; The final priority allocation table is used to obtain the data packet scheduling sequence of the network transport layer. The queue management technology is used to sort the data packets, determine the data packet processing order of the transport layer, and obtain the scheduling execution sequence.
[0015] S102. Dynamically adjust the network transmission queue in a high-concurrency environment according to the priority allocation table. If the data packet is identified as a critical type, immediately increase its transmission order and determine the adjusted transmission sequence.
[0016] The traffic information of the data packet is captured through the network transport layer, and the type identification field is extracted using data capture technology. The data packet is judged to belong to the key type or the secondary type based on the preset threshold parameters to obtain the initial classification set; For the initial classification set, real-time traffic data in a high-concurrency environment is obtained. Traffic characteristic values are calculated through statistical analysis. If the traffic characteristic value exceeds the dynamic threshold, the key type of data packets are marked as objects to be adjusted, and the set to be adjusted is obtained; Extract identification information of key types of data packets from the set to be adjusted, use the random forest algorithm to train the data packet features, and determine the adjustment range of the transmission sequence based on the feature weights to obtain the sequence adjustment parameters. According to the sequence adjustment parameters, the sequence status of the current transmission queue is obtained, and the transmission sequence of key type data packets is improved by using queue adjustment technology to determine the temporary transmission sequence; For the temporary transmission sequence, the traffic distribution characteristics of the data packets are obtained, the data packets are grouped using the K-means clustering algorithm, the stability of the transmission order after grouping is determined, and the optimized grouping set is obtained; Extract transmission sequence information from the optimized grouping set, use the decision tree algorithm to verify the grouping results, judge the rationality of the priority allocation of key types of data packets through the tree node splitting conditions, and determine the transmission sequence after verification; According to the verified transmission sequence, the scheduling queue status of the network transmission layer is obtained, the queue management technology is used to sort the data packets, and the final transmission sequence is determined by the sorting results.
[0017] S103 , extracting key data packets from the adjusted transmission sequence, transmitting them to the calculation processing layer, and matching the corresponding gesture parsing model through pre-established mapping rules to obtain initial aiming trajectory data.
[0018] The key data packet identification information is obtained through the adjusted transmission sequence, and the support vector machine algorithm is used to classify and train the data packet features to obtain a classification result set; Extract the traffic distribution characteristics of key data packets from the classification result set, group the data packets using the K-means clustering algorithm to obtain the grouped data set; Obtain queue status data for the grouped data set, use the decision tree algorithm to verify the priority of the grouping results, determine the priority allocation order based on the tree node splitting conditions, and obtain the verified priority set; Extracting feature weight information of key data packets from the verified priority set, calculating transmission sequence adjustment parameters based on the feature weights, and obtaining an adjusted sequence parameter set; The data input status of the computing processing layer is obtained according to the adjusted sequence parameter set, and the key data packets are sorted using queue management technology to obtain a sorted data sequence; Extracting matching information of the gesture parsing model from the sorted data sequence, performing model matching processing on the data sequence using a preset mapping rule, and obtaining a matched parsing data set; The initial distribution characteristics of the aiming trajectory are obtained for the matched analytical data set, and the trajectory data are optimized and adjusted through a linear regression algorithm to obtain the optimized aiming trajectory data.
[0019] S104: Obtain the current resource scheduling status in the computing processing layer. If the resource occupancy exceeds a preset threshold when a processing request for a critical data packet arrives, the dynamic allocation of the backup computing nodes is triggered to determine an optimized resource allocation plan.
[0020] Obtain resource scheduling status data of the computing processing layer, and use the occupancy detection module to determine whether the occupancy exceeds the preset threshold. If the occupancy exceeds the preset threshold, extract the key identifier of the data packet from the processing request queue to obtain the data set to be processed; For the data set to be processed, the availability of the backup nodes is sorted by the allocation algorithm to obtain a sorted node sequence; Obtain the scheduling status data of the standby node from the sorted node sequence, use the random forest algorithm to classify and train the node performance characteristics, and obtain a set of classified performance parameters; According to the classified performance parameter set, the load distribution order of the backup nodes is adjusted by a dynamic allocation technology to obtain an adjusted load parameter set; Based on the adjusted load parameter set, the real-time resource usage data of the computing processing layer is obtained, and queue management technology is used to prioritize the key processing requests of the data packets to obtain the sorted request sequence; Extract the resource allocation demand characteristics from the sorted request sequence, group the resource demands using the K-means clustering algorithm, and obtain the grouped demand set. According to the grouped demand set, the resource scheduling status data is optimized and adjusted through the linear regression algorithm to determine the final resource allocation plan.
[0021] S105: According to the optimized resource allocation plan, the initial aiming trajectory data is input into the allocated computing nodes, and the data is accelerated through the convolutional neural network algorithm to generate accurate aiming trajectory data.
[0022] By obtaining the optimized resource allocation scheme data, the initial aiming trajectory data is allocated to the target node in the computing node allocation using the data transmission process to obtain the transmitted trajectory data set; For the transmitted trajectory data set, the trajectory data features are extracted and processed through the convolutional neural network algorithm to obtain a feature-enhanced trajectory data set; Obtain a feature-enhanced trajectory data set, use accelerated processing technology combined with node computing power to parallel process the data, and obtain a preliminary processed trajectory data set; Extract real-time data scheduling information from the preliminarily processed trajectory data set, classify the scheduling priorities using the support vector machine algorithm, and obtain the classified priority sequence; According to the classified priority sequence, the real-time data scheduling technology is used to dynamically adjust the load in the computing node allocation to obtain the adjusted node load data; Based on the adjusted node load data, the algorithm processing performance is grouped and optimized using the K-means clustering algorithm to obtain a grouped performance parameter set; The generation requirement characteristics of precise aiming trajectories are obtained from the grouped performance parameter sets, and the linear regression algorithm is used to adjust the node computing power to determine the final trajectory generation data.
[0023] S106: extracting a timestamp sequence from the precise aiming trajectory data, and determining whether the data generation interval is less than the refresh period of the display rendering layer; if so, supplementing the intermediate frame data through an interpolation algorithm to obtain a continuous trajectory data set.
[0024] By obtaining a timestamp sequence from the precise aiming trajectory data, the technique of calculating the difference between adjacent timestamps is used to determine whether the data generation interval meets the refresh cycle requirements, and the interval judgment result is obtained; For the interval judgment result, if it is greater than the refresh period, the intermediate frame data is generated through the interpolation algorithm to obtain the supplemented trajectory data set; Obtain continuous trajectory features from the supplemented trajectory data set, use the random forest algorithm to classify the data point distribution, and obtain the classified trajectory point sequence; According to the classified trajectory point sequence, the smoothness parameter of the continuous trajectory is determined by calculating the spatial distance of each trajectory point, and the trajectory data after smoothness adjustment is obtained; For the trajectory data after smoothness adjustment, the K-nearest neighbor algorithm is used to cluster the coherence between trajectory points to obtain the clustered trajectory segment set; Obtain the correspondence between the timestamp sequence and the trajectory segment from the clustered trajectory segment set, predict the generation rate of the trajectory segment through the linear regression algorithm, and obtain the predicted rate data; According to the predicted rate data, a dynamic adjustment technology is used to optimize the refresh cycle of the display rendering layer to obtain an adjusted rendering parameter set.
[0025] S107. Based on the continuous trajectory data set, a real-time rendering pipeline is used to convert the data into visual feedback frames in the display rendering layer, and it is determined whether the time difference between the feedback frame sequence and the player's gesture input is lower than a preset threshold to obtain a synchronously optimized frame sequence.
[0026] By obtaining a timestamp sequence from a continuous trajectory dataset, processing the data using a real-time rendering pipeline to generate a visual feedback frame sequence, calculating the difference between the generation time of each feedback frame and the gesture input time, and determining whether the difference is below a preset threshold, a preliminary synchronized frame sequence set is obtained; For the initially synchronized frame sequence set, the correspondence between the timestamp sequence of the frame sequence and the continuous trajectory is obtained, and the time distribution features of the frame sequence are extracted through a convolutional neural network to obtain feature-enhanced frame sequence data; Extract the generated time series of visual feedback from the frame sequence data after feature enhancement, use the support vector machine algorithm to classify the fluctuation range of the time difference value, and obtain the classified time difference value set; According to the classified time difference value set, the correlation feature between each time difference value and the gesture input time is obtained, and the adjustment parameters for synchronization optimization are determined by calculating the mean and variance of the time difference values to obtain the adjusted synchronization parameter set; Based on the adjusted synchronization parameter set, the dynamic change characteristics of the continuous trajectory are obtained from the frame sequence set. The K-means clustering algorithm is used to cluster the timestamp sequence of the frame sequence to obtain the clustered frame segment set. By extracting the generation rate sequence of visual feedback from the clustered frame segment set, the time interval between the frame segments is smoothed using linear interpolation technology to obtain smoothed frame sequence data; Based on the smoothed frame sequence data, the processing delay characteristics of the rendering pipeline are obtained, and the frame generation frequency of real-time rendering is optimized through dynamic adjustment technology to obtain an optimized rendering parameter set.
[0027] S108: Obtain the synchronized and optimized frame sequence, update the screen content through the refresh mechanism of the head display device, and determine the final screen output data.
[0028] By acquiring the synchronized and optimized frame sequence, the refresh mechanism of the head-mounted display device is used to process the data, update the screen content, and obtain preliminary screen output data; Extract time series features from preliminary image output data, use convolutional neural networks to analyze the dynamic changes of frame sequences, and obtain feature-enhanced output data; For the output data after feature enhancement, the distribution characteristics of rendering frequency are obtained, and the processing rhythm of the refresh mechanism is optimized through dynamic adjustment technology to determine the adjusted rendering parameters; According to the adjusted rendering parameters, a smoothed interval sequence is extracted from the frame sequence, and the transition of the picture content is optimized using linear interpolation technology to obtain smoothed picture data; The smoothed image data is used to obtain the delay characteristics of the refresh mechanism of the head-mounted display device. The support vector machine algorithm is used to classify the delay range and determine the classified delay set. For the classified delay set, dynamically adjusted frequency features are extracted from the synchronized optimized frame sequence, and the distribution of the time series is clustered using the K-means clustering algorithm to obtain clustered frame segment data. Based on the clustered frame segment data, the generation rate sequence of the picture content is obtained, and the fluctuation of the rendering frequency is smoothed using dynamic adjustment technology to determine the final output data.
[0029] S109: Extract the player's expected matching index from the final screen output data. If the matching index is lower than the preset threshold, adjust the priority allocation table and resource scheduling plan through the reinforcement learning algorithm to generate the next round of optimization parameters.
[0030] The matching index data is extracted from the final screen output data, and the matching index data is classified and judged using a preset threshold. If the matching index is lower than the preset threshold, the priority allocation table and resource scheduling plan are adjusted through the reinforcement learning algorithm to obtain the adjusted optimization parameter data; Extract the time series distribution characteristics from the adjusted optimization parameter data, use dynamic adjustment technology to optimize the rhythm of the time series distribution characteristics, and determine the time series data after rhythm optimization. Obtain frequency distribution characteristic data from the time series data after rhythm optimization, and use smoothing technology to perform transition optimization on the frequency distribution characteristic data to obtain the frequency distribution data after transition optimization. Extracting a delay feature distribution sequence from the frequency distribution data after transition optimization, performing range classification on the delay feature distribution sequence using a support vector machine algorithm, and determining the classified delay feature data; The dynamic adjustment frequency sequence is obtained through the classified delay feature data, and the K-means clustering algorithm is used to perform distribution clustering on the dynamic adjustment frequency sequence to obtain the frequency distribution data after clustering; Extracting the picture generation rate sequence from the clustered frequency distribution data, using dynamic adjustment technology to perform fluctuation smoothing on the picture generation rate sequence, and determining the smoothed rate data; The optimized parameter adjustment sequence is obtained through the smoothed rate data, and the optimized parameter adjustment sequence is iteratively optimized using the reinforcement learning algorithm to obtain the final parameter data after iterative optimization.
[0031] In this embodiment, the type identification of each data packet in the network transmission layer is obtained through data capture technology, and the data packets are divided into key data packets (such as aiming instructions) and secondary data packets (such as ambient light and shadow synchronization) in combination with preset threshold parameters. The support vector machine algorithm is used to extract features and train classification boundaries for the data packet types to further optimize the classification results. According to the ratio of the number of classified data packets, the priority weight is calculated, and the priority of the key data packets is dynamically adjusted in combination with the real-time traffic characteristics. Finally, the stability of the priority allocation is verified by the decision tree algorithm, and the final priority allocation table is generated. On this basis, the queue management technology is used to dynamically adjust the transmission queue in a high-concurrency environment, give priority to key data packets, and ensure their rapid transmission; Extract key data packets from the adjusted transmission sequence and transmit them to the computing and processing layer. Generate initial aiming trajectory data by matching the corresponding gesture parsing model through preset mapping rules. During this process, monitor the resource occupancy status of the computing and processing layer in real time. When the processing request of the key data packet arrives and the resource occupancy exceeds the preset threshold, the dynamic allocation of the backup computing node is triggered. Use the random forest algorithm to classify and train the performance characteristics of the backup nodes, and adjust the load distribution order through dynamic allocation technology to optimize the resource scheduling plan. Subsequently, input the initial aiming trajectory data into the allocated computing node, and use the convolutional neural network algorithm to accelerate the data processing to generate accurate aiming trajectory data, further improving processing efficiency. A timestamp sequence is extracted from the precise aiming trajectory data to determine whether the data generation interval is less than the refresh period of the display rendering layer. If the interval is greater than the refresh period, an interpolation algorithm is used to supplement the intermediate frame data to generate a continuous trajectory dataset. Furthermore, a real-time rendering pipeline is used to convert the continuous trajectory data into visual feedback frames. A convolutional neural network and a K-means clustering algorithm are used to optimize the temporal distribution characteristics of the frame sequence, ensuring that the time difference between the feedback frame sequence and the player's gesture input is below a preset threshold. Finally, the refresh mechanism of the headset device updates the screen content, generating smooth and coherent visual feedback and enhancing the player's immersion.
[0032] The above is only a preferred embodiment of the present invention. It should be pointed out that ordinary technicians in this technical field can make several improvements and supplements without departing from the principles of the present invention. These improvements and supplements should also be regarded as the scope of protection of the present invention.
Claims
1. A method for reducing VR game interaction delay based on 5G, characterized in that: The method comprises: Obtain the type identifier of each data packet in the network transport layer, determine whether it is a key data packet or a secondary data packet through a preset threshold, and obtain a priority allocation table; According to the priority allocation table, the network transmission queue is dynamically adjusted in a high-concurrency environment. If a data packet is identified as a critical type, its transmission order is immediately promoted and the adjusted transmission sequence is determined; Extract key data packets from the adjusted transmission sequence and transmit them to the computing processing layer. Then, the corresponding gesture parsing model is matched with the pre-established mapping rules to obtain the initial aiming trajectory data. Obtain the current resource scheduling status in the computing processing layer. If the resource usage exceeds the preset threshold when the processing request of the critical data packet arrives, the dynamic allocation of the backup computing node is triggered and the optimized resource allocation plan is determined; According to the optimized resource allocation plan, the initial aiming trajectory data is input into the assigned computing nodes, the data is accelerated and processed to generate accurate aiming trajectory data; Extract the timestamp sequence from the precise aiming trajectory data and determine whether the data generation interval is less than the refresh period of the display rendering layer. If it is greater, interpolation algorithm is used to supplement the intermediate frame data to obtain a continuous trajectory data set. Based on the continuous trajectory dataset, a real-time rendering pipeline is used to convert the data into visual feedback frames in the display rendering layer. The time difference between the feedback frame sequence and the player's gesture input is determined to be below a preset threshold, and a synchronized and optimized frame sequence is obtained. Obtain the synchronized and optimized frame sequence, update the screen content through the refresh mechanism of the head-mounted display device, and determine the final screen output data; The player's expected matching index is extracted from the final screen output data. If the matching degree is lower than the preset threshold, the priority allocation table and resource scheduling plan are adjusted to generate the next round of optimization parameters.
2. The method according to claim 1, characterized in that Obtaining the priority allocation table specifically includes: Obtain the type identification of each data packet from the network transport layer, extract the identification field using data capture technology, determine the original type information of the data packet, and obtain a type identification set; For the type identification set, a preset threshold parameter is obtained, and by comparing the type identification value with the preset threshold, each data packet is judged to be a key data packet or a secondary data packet, and a preliminary classification result is obtained; Based on the preliminary classification results, the data packet type is characterized and the classification boundary is trained through the feature vector to determine the precise division between key data packets and secondary data packets, thus obtaining the optimized classification set. Extract the ratio of key data packets to secondary data packets from the optimized classification set, calculate the priority weight through statistical analysis, determine the initial value of priority allocation, and obtain a weight allocation table; According to the weight allocation table, the real-time traffic characteristics of the data packets are obtained. If the traffic characteristics exceed the preset dynamic threshold, the priority value of the key data packet is adjusted to obtain a dynamic priority adjustment table; According to the dynamic priority adjustment table, the data packet classification is verified twice, and the stability of the priority allocation is judged by the splitting conditions of the tree nodes to obtain the final priority allocation table; The final priority allocation table is used to obtain the data packet scheduling sequence of the network transport layer. The queue management technology is used to sort the data packets, determine the data packet processing order of the transport layer, and obtain the scheduling execution sequence.
3. The method according to claim 1, characterized in that Determining the adjusted transmission sequence specifically includes: The traffic information of the data packet is captured through the network transport layer, and the type identification field is extracted using data capture technology. The data packet is judged to belong to the key type or the secondary type based on the preset threshold parameters to obtain the initial classification set; For the initial classification set, real-time traffic data in a high-concurrency environment is obtained. Traffic characteristic values are calculated through statistical analysis. If the traffic characteristic value exceeds the dynamic threshold, the key type of data packets are marked as objects to be adjusted, and the set to be adjusted is obtained; Extract identification information of key types of data packets from the set to be adjusted, train the data packet features, determine the adjustment range of the transmission sequence through feature weights, and obtain the sequence adjustment parameters; According to the sequence adjustment parameters, the sequence status of the current transmission queue is obtained, and the transmission sequence of key type data packets is improved by using queue adjustment technology to determine the temporary transmission sequence; For the temporary transmission sequence, the traffic distribution characteristics of the data packets are obtained, the data packets are grouped, the stability of the transmission order after grouping is determined, and the optimized grouping set is obtained; Extract transmission order information from the optimized grouping set, verify the grouping results, judge the rationality of priority allocation of key type data packets through tree node splitting conditions, and determine the transmission sequence after verification; According to the verified transmission sequence, the scheduling queue status of the network transmission layer is obtained, the queue management technology is used to sort the data packets, and the final transmission sequence is determined by the sorting results.
4. The method according to claim 1, wherein The obtaining of the initial aiming trajectory data specifically includes: Obtain key data packet identification information through the adjusted transmission sequence, perform classification training on the data packet features, and obtain a classification result set; Extract the traffic distribution characteristics of key data packets from the classification result set, group the data packets, and obtain the grouped data set; Obtain queue status data for the grouped data set, perform priority verification on the grouping results, determine the priority allocation order based on the tree node splitting conditions, and obtain the verified priority set; Extracting feature weight information of key data packets from the verified priority set, calculating transmission sequence adjustment parameters based on the feature weights, and obtaining an adjusted sequence parameter set; The data input status of the computing processing layer is obtained according to the adjusted sequence parameter set, and the key data packets are sorted using queue management technology to obtain a sorted data sequence; Extracting matching information of the gesture parsing model from the sorted data sequence, performing model matching processing on the data sequence using a preset mapping rule, and obtaining a matched parsing data set; The initial distribution characteristics of the aiming trajectory are obtained for the matched analytical data set, and the trajectory data are optimized and adjusted through a linear regression algorithm to obtain the optimized aiming trajectory data.
5. The method according to claim 1, characterized in that Determining the optimized resource allocation plan specifically includes: Obtain resource scheduling status data of the computing processing layer, and use the occupancy detection module to determine whether the occupancy exceeds the preset threshold. If the occupancy exceeds the preset threshold, extract the key identifier of the data packet from the processing request queue to obtain the data set to be processed; For the data set to be processed, the availability of the backup nodes is sorted to obtain a sorted node sequence; Obtain the scheduling status data of the standby node from the sorted node sequence, perform classification training on the node performance characteristics, and obtain a set of classified performance parameters; According to the classified performance parameter set, the load distribution order of the backup nodes is adjusted by a dynamic allocation technology to obtain an adjusted load parameter set; Based on the adjusted load parameter set, the real-time resource usage data of the computing processing layer is obtained, and queue management technology is used to prioritize the key processing requests of the data packets to obtain the sorted request sequence; Extract the resource allocation demand characteristics from the sorted request sequence, group the resource demands, and obtain the grouped demand set; Based on the grouped demand sets, the resource scheduling status data is optimized and adjusted to determine the final resource allocation plan.
6. The method according to claim 1, characterized in that The precise aiming trajectory data specifically includes: By obtaining the optimized resource allocation scheme data, the initial aiming trajectory data is allocated to the target node in the computing node allocation using the data transmission process to obtain the transmitted trajectory data set; For the transmitted trajectory data set, the trajectory data features are extracted and processed to obtain a feature-enhanced trajectory data set; Obtain a feature-enhanced trajectory data set, use accelerated processing technology combined with node computing power to parallel process the data, and obtain a preliminary processed trajectory data set; Extracting real-time data scheduling information from the preliminarily processed trajectory data set, classifying the scheduling priorities, and obtaining a classified priority sequence; According to the classified priority sequence, the real-time data scheduling technology is used to dynamically adjust the load in the computing node allocation to obtain the adjusted node load data; Based on the adjusted node load data, the algorithm processing performance is optimized in groups to obtain a grouped performance parameter set; The generation requirement characteristics of precise aiming trajectories are obtained from the grouped performance parameter sets, the node computing power is adjusted, and the final trajectory generation data is determined.
7. The method according to claim 1, characterized in that The obtaining of the continuous trajectory dataset specifically includes: By obtaining a timestamp sequence from the precise aiming trajectory data, the technique of calculating the difference between adjacent timestamps is used to determine whether the data generation interval meets the refresh cycle requirements, and the interval judgment result is obtained; For the interval judgment result, if it is greater than the refresh period, the intermediate frame data is generated through the interpolation algorithm to obtain the supplemented trajectory data set; Obtain continuous trajectory features from the supplemented trajectory data set, classify the data point distribution, and obtain a classified trajectory point sequence; According to the classified trajectory point sequence, the smoothness parameter of the continuous trajectory is determined by calculating the spatial distance of each trajectory point, and the trajectory data after smoothness adjustment is obtained; For the trajectory data after smoothness adjustment, clustering is performed on the coherence between trajectory points to obtain a clustered trajectory segment set; Obtain the correspondence between the timestamp sequence and the trajectory segment from the clustered trajectory segment set, predict the generation rate of the trajectory segment, and obtain the predicted rate data; According to the predicted rate data, a dynamic adjustment technology is used to optimize the refresh cycle of the display rendering layer to obtain an adjusted rendering parameter set.
8. The method according to claim 1, characterized in that The obtaining of the frame sequence after synchronization optimization specifically includes: By obtaining a timestamp sequence from a continuous trajectory dataset, processing the data using a real-time rendering pipeline to generate a visual feedback frame sequence, calculating the difference between the generation time of each feedback frame and the gesture input time, and determining whether the difference is below a preset threshold, a preliminary synchronized frame sequence set is obtained; For the initially synchronized frame sequence set, the correspondence between the timestamp sequence of the frame sequence and the continuous trajectory is obtained, and the time distribution features of the frame sequence are extracted through a convolutional neural network to obtain feature-enhanced frame sequence data; Extracting the generated time series of visual feedback from the feature-enhanced frame sequence data, classifying the fluctuation range of the time difference, and obtaining a classified time difference set; According to the classified time difference value set, the correlation feature between each time difference value and the gesture input time is obtained, and the adjustment parameters for synchronization optimization are determined by calculating the mean and variance of the time difference values to obtain the adjusted synchronization parameter set; For the adjusted synchronization parameter set, the dynamic change characteristics of the continuous trajectory are obtained from the frame sequence set, and the timestamp sequence of the frame sequence is clustered to obtain the clustered frame segment set; By extracting the generation rate sequence of visual feedback from the clustered frame segment set, the time interval between the frame segments is smoothed using linear interpolation technology to obtain smoothed frame sequence data; Based on the smoothed frame sequence data, the processing delay characteristics of the rendering pipeline are obtained, and the frame generation frequency of real-time rendering is optimized through dynamic adjustment technology to obtain an optimized rendering parameter set.
9. The method according to claim 1, characterized in that The determining of the final screen output data specifically includes: By acquiring the synchronized and optimized frame sequence, the refresh mechanism of the head-mounted display device is used to process the data, update the screen content, and obtain preliminary screen output data; Extract time series features from preliminary image output data, use convolutional neural networks to analyze the dynamic changes of frame sequences, and obtain feature-enhanced output data; For the output data after feature enhancement, the distribution characteristics of rendering frequency are obtained, and the processing rhythm of the refresh mechanism is optimized through dynamic adjustment technology to determine the adjusted rendering parameters; According to the adjusted rendering parameters, a smoothed interval sequence is extracted from the frame sequence, and the transition of the picture content is optimized using linear interpolation technology to obtain smoothed picture data; Obtain the delay characteristics of the refresh mechanism of the head-mounted display device through the smoothed image data, classify the delay range, and determine the classified delay set; For the classified delay set, dynamically adjusted frequency features are extracted from the synchronized and optimized frame sequence, and the distribution of the time series is clustered to obtain clustered frame segment data. Based on the clustered frame segment data, the generation rate sequence of the picture content is obtained, and the fluctuation of the rendering frequency is smoothed using dynamic adjustment technology to determine the final output data.
10. The method according to claim 1, characterized in that The generation of the next round of optimization parameters specifically includes: The matching index data is extracted from the final screen output data, and the matching index data is classified and judged using a preset threshold. If the matching index is lower than the preset threshold, the priority allocation table and resource scheduling plan are adjusted through the reinforcement learning algorithm to obtain the adjusted optimization parameter data; Extracting time series distribution features from the adjusted optimization parameter data, optimizing the rhythm of the time series distribution features using dynamic adjustment technology, and determining the time series data after rhythm optimization; Obtain frequency distribution characteristic data through the time series data after rhythm optimization, and use smoothing technology to perform transition optimization on the frequency distribution characteristic data to obtain transition optimized frequency distribution data; extracting a delay feature distribution sequence from the frequency distribution data after transition optimization, performing range classification on the delay feature distribution sequence, and determining the classified delay feature data; Obtaining a dynamically adjusted frequency sequence through the classified delay feature data, performing distribution clustering on the dynamically adjusted frequency sequence, and obtaining frequency distribution data after clustering; Extracting the picture generation rate sequence from the clustered frequency distribution data, using dynamic adjustment technology to perform fluctuation smoothing on the picture generation rate sequence, and determining the smoothed rate data; The optimized parameter adjustment sequence is obtained through the smoothed rate data, and the optimized parameter adjustment sequence is iteratively optimized using the reinforcement learning algorithm to obtain the final parameter data after iterative optimization.